20230041233. IMAGE RECOGNITION METHOD AND APPARATUS, COMPUTING DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM simplified abstract (Tencent Technology (Shenzhen) Company Limited)
Contents
IMAGE RECOGNITION METHOD AND APPARATUS, COMPUTING DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM
Organization Name
Tencent Technology (Shenzhen) Company Limited
Inventor(s)
IMAGE RECOGNITION METHOD AND APPARATUS, COMPUTING DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM - A simplified explanation of the abstract
This abstract first appeared for US patent application 20230041233 titled 'IMAGE RECOGNITION METHOD AND APPARATUS, COMPUTING DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM
Simplified Explanation
The abstract describes an image recognition method that uses a generative adversarial network (GAN) to determine if an image is forged or not. The GAN consists of a generator and a classifier. The classifier is trained using a group of original images, each with a category label. The method involves obtaining first-type noise corresponding to each original image, inputting the original image into the generator to obtain an output, and obtaining second-type noise corresponding to the original image as the output. The classifier is then trained using the original image, first-type noise, and second-type noise.
- Image recognition method using a generative adversarial network (GAN)
- GAN includes a generator and a classifier
- Classifier is trained using a group of original images and their category labels
- First-type noise is obtained for each original image
- Original image is inputted into the generator to obtain an output
- Second-type noise is obtained corresponding to the original image as the output
- Classifier is trained using the original image, first-type noise, and second-type noise
Potential Applications
- Image forgery detection
- Enhancing security in digital media
- Authenticating images in various industries such as journalism, forensics, and e-commerce
Problems Solved
- Detecting forged images with high accuracy
- Improving the reliability of image recognition systems
- Addressing the growing concern of image manipulation and forgery
Benefits
- Provides a method to determine if an image is forged or not
- Utilizes a generative adversarial network for improved accuracy
- Trains the classifier using original images and corresponding noise
- Can be applied in various industries to ensure the authenticity of images
Original Abstract Submitted
an image recognition method includes: obtaining a to-be-recognized image; determining whether the image is a forged image by recognizing the image through a trained generative adversarial network, the generative adversarial network including a generator and a classifier. training the classifier includes: obtaining an original image group having a plurality of original images, and a category label of each original image. each of the plurality of original images includes a real image and a forged image corresponding to the real image. the method includes obtaining using the classifier, for a respective original image of the plurality of original images, first-type noise corresponding to the respective original image; inputting the respective original image into the generator to obtain an output of the generator, and obtaining second-type noise corresponding to the respective original image as the output; and training the classifier using the respective original image, the first-type noise, and the second-type noise.
